[Bug] Opus 5.5 incorrectly flags CVP-approved cybersecurity experts with [cyber] safeguard block
Bug Description
Hi, I'm a cybersecurity expert approved in CVP but still flaged [cyber] and blocked from opus 5.5 , even with a new session and a single "hello", trace:
❯ salut
● Opus 5.5's safeguards stopped the response above · continuing once with that noted
────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
Session paused
Opus 5.5's safeguards flagged this session. You may be seeing this for the first time: Opus 5.5 is more capable and has stronger safeguards as a result, which can sometimes flag non-cybersecurity work. We're improving these safeguards to reduce the amount of incorrectly flagged messages. Edit and retry, or continue with Opus 4.8. Send feedback with /feedback or learn more
Details: [cyber]
❯ 1. Switch to Opus 4.8
2. Edit prompt and retry with Opus 5.5
Environment Info
- Platform: linux
- Terminal: WarpTerminal
- Version: 2.1.284
- Feedback ID: cf5beb58-e6a5-4674-bea3-ef39a7f8c0b2
Errors
[]This issue has 1 comment on GitHub. Read the full discussion on GitHub ↗